Schizophrenia with onset in later life shows important similarities and differences compared to more typical cases with early adult life onset. The investigator proposes to study cerebral structure and dopamine D2 receptors in late life onset schizophrenics and to compare these with two individually matched control groups; 1) currently elderly, EOS, and 2) elderly normal volunteers. He will use MRI and PET scanning techniques initially developed by the investigators to study early-onset cases; MRI will be used to explore differences in heteromodal association neocortical regions, and PET scans, using C-11-labeled NMSP as the ligand, will measure the Bmax values of dopamine D2 receptors. The investigator expects to see many of the same changes as in early onset patients, once age-related confounds are removed. Brain structural and receptor measures will be analyzed to determine their (covariance) association with carefully characterized clinical symptomatology, cognitive abnormalities, sensory deficits and social factors. This may aid in the understanding and subtyping of late life onset schizophrenia, and help clarify the relationship of this syndrome both to the early onset illness and to the psychopathology of aging and of psychosis. A five-year project is proposed in order to collect a sufficient number of drug-naive index patients, and to enable follow-up to confirm diagnosis. Because the rate-limiting step is the number of drug-naive late onset schizophrenics, the investigator has elected to study these index subjects in considerable detail. Significant time needs to be spent in screening potential participants, contacting clinics, impatient services and the psychiatric emergency room, eliminating individuals with exclusion factors, and scheduling and interpreting studies. The laboratory has an effective infrastructure for carrying out and interpreting a variety of clinical, imaging nad physiologic studies. Key equipment and sophisticated quantitative graphic methods are in place; funds are therefore requested mainly for scans and for personnel. The investigator's last budget underestimated the personnel costs of getting the job done. Now that he has the staff, equipment and novel quantitative procedures, he can benefit from economies of scale and study more relevant comparison subjects, to better address key hypotheses.